Beruflich Dokumente
Kultur Dokumente
PUBLIC LAW AND LEGAL THEORY WORKING PAPER NO. 361
LATIN AMERICAN PRESIDENTIALISM IN COMPARATIVE AND
HISTORICAL PERSPECTIVE
José Antonio Cheibub, Zachary Elkins, and Tom Ginsburg
THE LAW SCHOOL
THE UNIVERSITY OF CHICAGO
September 2011
This paper can be downloaded without charge at the Public Law and Legal Theory Working Paper
Series: http://www.law.uchicago.edu/academics/publiclaw/index.html
and The Social Science Research Network Electronic Paper Collection.
note 9, at 531 (―Parliamentarism and presidentialism are commonly, and correctly, set in opposition
as distinguishable systems of governance that exhibit distinguishable structural features.‖).
13. See JOSÉ ANTONIO CHEIBUB, PRESIDENTIALISM, PARLIAMENTARISM, AND DEMOCRACY
36–37 (2007) (defining assembly confidence as a political system in which the government‘s
authority is constrained by the continued approval and confidence of the legislative assembly).
14. See infra notes 32–42 and accompanying text.
15. See Keith S. Rosenn, Separation of Powers in Brazil, 47 DUQ. L. REV. 839, 840–42 (2009)
(describing the legislative power under the 1824 constitution as parliamentary and noting that the
1824 constitution was Brazil‘s ―most enduring‖).
16. Donald L. Horowitz, The Federalist Abroad in the World, in THE FEDERALIST PAPERS 502,
505 (Ian Shapiro ed., 2009); Robert J. Kolesar, North American Constitutionalism and Spanish
America: “A Special Lock Ordered by Catalogue, Which Arrived with the Wrong Instructions and
No Keys”?, in AMERICAN CONSTITUTIONALISM ABROAD 41, 53–54 (George Athan Billias ed.,
1990); Miguel Schor, Constitutionalism Through the Looking Glass of Latin America, 41 TEX.
INT‘L L.J. 1, 15 (2006).
17. Kolesar, supra note 16, at 53–56. For a thorough discussion of this topic, see
JONATHAN M. MILLER, BORROWING A CONSTITUTION: THE U.S. CONSTITUTION IN ARGENTINA
AND THE HEYDAY OF THE ARGENTINE SUPREME COURT (1853–1930) (forthcoming 2012).
18. Horowitz, supra note 16, at 505 (quoting BERNARD BAILYN, TO BEGIN THE WORLD ANEW
146 (2003)).
19. Kolesar, supra note 16, at 42–43; see also Zachary Elkins, Diffusion and the
Constitutionalization of Europe, 43 COMP. POL. STUD. 969, 984 (2010) (comparing the influence of
different constitutional models on constitutions in Europe to that process in Latin America).
20. Kolesar, supra note 16, at 43–44.
21. Id. at 43.
22. See id. at 51 (―[D]uring the early years of independence, . . . North American constitutional
principles came to be closely associated with federalism in Chile.‖).
23. Horowitz, supra note 16, at 505. For background on the Central American Federation, see
LYNN V. FOSTER, A BRIEF HISTORY OF CENTRAL AMERICA 134–51 (2000).
24. See DAVID BUSHNELL, THE MAKING OF MODERN COLOMBIA: A NATION IN SPITE OF
ITSELF 51–52 (1993) (describing the process by which Gran Colombia became a federal republic).
25. ZACHARY ELKINS ET AL., THE ENDURANCE OF NATIONAL CONSTITUTIONS 26 (2009).
2011] Latin American Presidentialism 5
26. See Keith S. Rosenn, Judicial Review in Latin America, 31 OHIO ST. L.J. 785, 785 (1974)
(―A region of chronic political instability and short-lived constitutions with a civil law tradition
would appear most infertile soil for the seeds of Marbury v. Madison to take root. Yet all of the
Latin American republics, with the exception of the Dominican Republic, provide for some form of
judicial review.‖ (footnotes omitted) (citing Marbury v. Madison, 5 U.S. (1 Cranch) 137 (1803))).
27. Id. at 796.
28. See id. at 791–808 (surveying the historical development of mechanisms for raising
constitutional questions within Latin American countries).
29. ELKINS ET AL., supra note 25, at 27 & fig.2.2.
30. Horowitz, supra note 16, at 505.
31. ELKINS ET AL., supra note 25, at 28.
32. CHEIBUB, supra note 13, at 150.
33. See Adam Przeworski et al., The Origins of Parliamentary Responsibility, in COMPARATIVE
CONSTITUTIONAL DESIGN (Tom Ginsburg ed., forthcoming 2011) (manuscript at 150, 156–61) (on
file with authors) (detailing the rise of parliamentary monarchies in Europe and tracing the ―shift of
the power to appoint governments from the crown to elected assemblies‖).
6 Texas Law Review [Vol. 89:XXXX
period of negotiation between the monarch and the nobles, in which the par-
liament ultimately displaced the monarchy as the center of effective
governance.34 In Latin America, by contrast, initial governments, whether
revolutionary or not, emerged from a system of monarchy in which a single
individual sat at the center of the political system.35 Even Simón Bolívar,
who cloaked his critique of federalism in a general reaction to borrowing
from North America, was an admirer of the presidency as a model of a
nonhereditary yet strong executive.36 Thus, the drafters of presidential
constitutions in nineteenth-century Latin America did not choose between the
presidential and parliamentary models available today, but between a monar-
chy (headed by a hereditary leader) and a republic (headed by leaders with no
claim to heredity).37
At that time, the United States represented the most successful republic
and one that had emerged under similar circumstances.38 France, after all,
had not yet been able to settle upon a stable and coherent model of republi-
can government.39 Adoption of a presidential formula was perhaps a
foregone conclusion.
At the same time, the initial choice of presidentialism does not
necessarily explain the endurance of that model. After all, many other
institutions were discarded over time through processes of amendment and
constitutional replacement. There are reasons, however, to suppose that
basic constitutional frameworks—such as the one embodied in the proce-
dures for the selection of the executive—are subject to strong inertial factors.
These broad institutions structure the expectations of the actors operating
under them and, in order for them to be changed, actors must be willing to
leap into the unknown. At the same time, constitutions serve as focal points
and are rarely written on a blank slate; previous documents often serve as a
template, even if changes are made to address issues identified as leading to
crisis in prior systems of government. Thus, in spite of frequent
34. See id. at 156–57 (noting that although ―[c]onstitutional monarchs were chief executives . . .
who governed with the advice and consent of their ministers[,] . . . there were many instances in
which parliamentary majorities forced monarchs to dismiss or accept governments against their
will‖ and that ―[t]he power of the parliaments stemmed from their control over legislation,
particularly budgets‖).
35. See id. at 175–76 (describing the development of constitutions in Portugal and Spain and
noting the central role of the monarch in each nation).
36. Kolesar, supra note 16, at 50.
37. CHEIBUB, supra note 13, at 151.
38. See Kolesar, supra note 16, at 44 (noting that the ―social and economic success of the
United States‖ prompted Latin American drafters to consider the principles embodied in the
Constitution); id. at 58 (―North American constitutionalism was influential precisely because it
embodied values and addressed needs shared by many [Latin] Americans.‖).
39. The first stable republican government in France emerged in 1875. See ELKINS ET AL.,
supra note 25, at 169 (―The constitution that emerged [in 1875] was a compromise that combined a
strong chamber of deputies elected by universal suffrage and an upper house composed of senators
selected by local notables or appointed for life terms. Combining both popular and conservative
impulses, these institutions nevertheless facilitated the dominance of republicanism . . . .‖).
2011] Latin American Presidentialism 7
and the executive (president) signs or vetoes them.‖44 Others emphasize the
following as key attributes of political systems: decree power,45 emergency
rule,46 veto power,47 legislative initiative,48 cabinet formation,49 and the
power to dissolve the assembly.50 This last feature is so closely linked with
parliamentarism that some even include it as a defining attribute.51
The United States Constitution represents the archetypical presidential
system in the sense that it is the model that represents, often implicitly,
discussions of separation-of-powers systems. What defines the U.S.
Constitution as presidential is that the executive is popularly elected and does
not need the confidence of the legislature in order to remain in office.52
Other features of the U.S. presidential system may or may not be unique and
include the following: First, the U.S. President is unable to dissolve the
assembly.53 Second, the President lacks explicit lawmaking powers and has
44. George Tsebelis, Decision Making in Political Systems: Veto Players in Presidentialism,
Parliamentarism, Multicameralism and Multipartyism, 25 BRIT. J. POL. SCI. 289, 325 (1995).
45. See, e.g., Lee Kendall Metcalf, Measuring Presidential Power, 33 COMP. POL. STUD. 660,
663 tbl.1 (2000) (citing Timothy Frye, A Politics of Institutional Choice: Post-Communist
Presidencies, 30 COMP. POL. STUD. 523 (1997)) (including the power to ―[i]ssue[] decrees in non
emergencies‖ among Frye‘s twenty-seven listed presidential powers).
46. See, e.g., BRIAN LOVEMAN, THE CONSTITUTION OF TYRANNY: REGIMES OF EXCEPTION IN
SPANISH AMERICA 5–6 (1993) (―Latin American constitutions almost always included provisions
for ‗emergency powers,‘ . . . to be used in times of internal strife or external threat.‖).
47. See, e.g., Albert, supra note 9, at 542–43 (characterizing the presidential veto in the U.S.
Constitution as a legislative power).
48. See, e.g., José Antonio Cheibub, Making Presidential and Semi-presidential Constitutions
Work, 87 TEXAS L. REV. 1375, 1386–88 (2009) (noting that ―[a]lmost all presidential constitutions
give some legislative powers to the presidency,‖ including the ―exclusive power to introduce
legislation in some specified areas‖).
49. See, e.g., Metcalf, supra note 45, at 660, 663 tbl.1 (citing Frye, supra note 45) (including
the power to ―[a]ppoint[] senior officers‖ among Frye‘s twenty-seven listed presidential powers).
50. See, e.g., Krouwel, supra note 9, at 339, 342–45 (distinguishing presidential, semi-
presidential, and parliamentary systems on several dimensions, including the ability of various
political actors to dissolve the legislature).
51. See, e.g., Stepan & Skach, supra note 12, at 3 (including the executive‘s ability to dissolve
the legislature as one of two ―fundamental characteristics‖ of a ―pure parliamentary regime‖).
52. See U.S. CONST. art. II, § 1, cl. 1–3, amended by U.S. CONST. amend. XII (providing for a
fixed presidential term of four years and popular election of the president through the electoral
college).
53. See U.S. CONST. art. II, amended by U.S. CONST. amend. XII & XXV (defining the powers
of the executive, which do not include the power to dissolve Congress). Although the power to
dissolve the assembly is often considered to be an essential, even defining, feature of the separation
of powers system, we do not take this position. Dissolution powers originated in monarchies and
are compatible today with all forms of democratic constitutions. Just as there are presidential
constitutions that allow dissolution under certain circumstances, there are parliamentary ones that
do not. Compare CONSTITUCIÓN DE LA REPÚBLICA DEL ECUADOR [ECUADOR CONST.] 2008,
art. 148 (listing certain circumstances under which the president can dissolve the national
assembly), with Przeworski et al., supra note 33, at 158 (noting Norway as an exception to the
general rule that, in countries operating under a constitutional monarchy, kings can dissolve
parliaments).
2011] Latin American Presidentialism 9
54. See U.S. CONST. art. I, § 1 (―All legislative Powers herein granted shall be vested in a
Congress of the United States, which shall consist of a Senate and House of Representatives.‖).
55. See infra note 78 and accompanying text.
56. See U.S. CONST. art. I, § 7, cl. 1–2 (stating how legislation may be introduced by the House
and the Senate, and laying out the President‘s veto power).
57. See Myers v. United States, 272 U.S. 52, 164 (1926) (holding that the President‘s power to
appoint officers entails the power to remove them, but that the Appointments Clause does not
require the Senate‘s consent to the removal).
58. See U.S. CONST. art. II, § 2, cl. 1 (―The President . . . shall have Power to grant Reprieves
and Pardons for Offences against the United States, except in Cases of Impeachment.‖).
59. Tom Ginsburg, James Melton & Zachary Elkins, On the Evasion of Executive Term Limits,
52 WM. & MARY L. REV. 1807, 1819, 1834–35 (2011) (discussing the development of informal
presidential term limits and the eventual ratification of the Twenty-Second Amendment).
60. See, e.g., U.S. CONST. art. I, § 2, cl. 5 (granting the House the power to impeach executive
officials); id. art. I, § 3, cl. 6 (granting the Senate the power to try all impeachment cases); id. art. II,
§ 2, cl. 2 (limiting the President‘s power to make treaties and appointments to those made ―with the
Advice and Consent of the Senate‖).
61. See Ginsburg, Melton & Elkins, supra note 59, at 1834–35 (explaining that George
Washington‘s service of only two terms led to the creation of an ―unwritten constitutional norm‖).
62. See WILLIAM G. HOWELL, POWER WITHOUT PERSUASION: THE POLITICS OF DIRECT
PRESIDENTIAL ACTION 84 & fig.4.1, app. at 189–91 (2003) (demonstrating the increase in
―significant executive orders‖ during the twentieth century across a diverse set of policy categories).
63. See Jide Nzelibe, A Positive Theory of the War Powers Constitution, 91 IOWA L. REV. 993,
996–97 & nn.2–4 (2006) (discussing the debate over executive powers as one between pro-
President scholars, who stress the importance of strength and flexibility in an executive, and pro-
Congress scholars, who argue that a legislative check on the President‘s foreign-policy power
encourages democratic accountability).
10 Texas Law Review [Vol. 89:XXXX
Appendix A provides some sense of the population and our sample. Our
sample includes 193 of the 231 systems, or 81%. The thirty-eight
constitutions missing from our sample tend to be concentrated in the early
years after independence.69 Of the 111 systems adopted (and discarded)
before 1880, our sample includes seventy-nine systems—roughly four-fifths
of that population. These early years are precisely the years of institutional
vacuum that followed independence, when there was the highest degree of
constitutional experimentation. That this population of systems is
underrepresented in our sample implies that our estimate of intraregional
diversity may be biased towards increased homogeneity in the first decades
of the nineteenth century.
We start by considering a set of thirteen attributes, ten pertaining to
powers allocated to the executive and three to powers allocated to the
legislature. Regarding the executive, we consider the following powers: to
issue executive decrees, to assume emergency powers, to propose constitu-
tional amendments, to propose the budget law, to initiate regular legislation,
to veto legislation, to issue pardons, to appoint and dismiss the cabinet, and
to dissolve the legislature. Regarding the legislature, we consider the
legislature‘s power to remove individual ministers, to exercise oversight over
the executive, and to override the executive veto (assuming the constitution
provides for such veto).
Appendix B presents the proportion of Latin American constitutions
with selected executive–legislative provisions over time. For the temporal
dimension, we divide the region‘s history into five eras: (1) a period of
economic and political disorganization (independence through 1870); (2) the
period of agro-export development, during which most countries in the
region were integrated into the international economy as exporters of raw
material and importers of industrialized goods (1870–1918); (3) the period of
crisis of the export model and emergence of import-substitution
industrialization (1919–1945); (4) the period of dominance and then decline
of import-substitution industrialization (1946–1979); and (5) the period of
democratization and economic reforms (1979–2007).
This table displays a remarkable evolution in executive powers across
Latin American constitutions. Let us start, however, with the less remarkable
features of the table. Given that most Latin American constitutions have
been presidential, it is not surprising that the number of constitutions that
allow presidents to dissolve the legislature is relatively small. This number,
however, is not trivial: overall, there have been seventeen Latin American
constitutions that allowed the executive to dissolve the legislature; of these,
eleven are classified as presidential. Equally unsurprising is the fact that
close to 90% of the constitutions written since independence have granted the
69. Of the thirty-eight cases not sampled, thirty-two are constitutions that were written before
1860.
12 Texas Law Review [Vol. 89:XXXX
executive the power to freely appoint and dismiss the cabinet. And virtually
every Latin American constitution grants the executive emergency power
(although there is considerable variation regarding the specifics of this
power, as we will see below). Finally, many constitutions have established
relatively strong legislatures, at least when it comes to oversight of the exec-
utive (a feature that has been almost universal since 1870), removal of
individual ministers (about one-half of all Latin American constitutions so
allow), and override of an executive veto (almost all of the post-World
War II constitutions provide for it).
The remarkable development, in our view, is the increase in provisions
that grant the executive some lawmaking powers. A high proportion of
executives have always been given decree powers in Latin American
constitutions, but twentieth-century constitutions rendered this provision
almost universal (although, again, there is considerable variation in the
specifics of this power, as we will see below). Equally prevalent has been
the executive‘s veto power: close to 90% for the whole period and universal
for the post-1979 period. But, whereas less than 10% of the constitutions
written in the nineteenth century allowed the executive to propose constitu-
tional amendments, the proportion in the post-1979 period has soared to
90%. Although less dramatic, a similar pattern is evident with respect to the
executive‘s capacity to initiate ordinary legislation and to propose budget
legislation.
Thus, we see some convergence in Latin American constitutions in the
sense of an expansion of the powers of the executive, particularly executive
lawmaking powers. At the same time, powers that were relatively common
in earlier constitutions either did not change much or have expanded in more
recent times. This pattern can be observed in Appendix C, which plots the
proportion of constitutions in force that provide for a given power. This
convergence includes the features normally associated with presidential
constitutions—the executive‘s power to appoint and dismiss the cabinet, and
the inability to dissolve the legislature. One preliminary observation may be
that the data suggest a contemporary pattern of Latin American constitution-
alism that combines a strong legislature with a president possessing strong
lawmaking powers. This contrasts with the earlier pattern of strong legisla-
tures with presidents possessing few or no lawmaking powers.70
How unique is this pattern with respect to other presidential
constitutions? Is the evolution of Latin American constitutions toward
broader legislative power for the executive a region-specific development, or
70. Interestingly, the earlier pattern has been identified by Shugart and Carey as a configuration
conducive to regime survival, while the current configuration is viewed by them as detrimental to
successful governments. MATTHEW SOBERG SHUGART & JOHN M. CAREY, PRESIDENTS AND
ASSEMBLIES 277 (1992). A test of this proposition is beyond the scope of this Article, but we note
that the earlier period was associated with instability in constitutional form. See supra text
accompanying note 69.
2011] Latin American Presidentialism 13
71. Before 1940, there were nine presidential constitutions outside of Latin America: the United
States (1789), Haiti (1843 and 1935), France (1848), Germany (1919), Lithuania (1938), Liberia
(1847), and the Philippines (1899 and 1935). Since 1940, there have been eighty presidential
constitutions written in countries outside of Latin America. Note that the 1919 German constitution
did not explicitly provide for a directly elected president, and for this reason it is not classified as a
semi-presidential constitution. For the classification of constitutions as presidential, parliamentary,
and semi-presidential, see generally Cheibub, Elkins & Ginsburg, supra note 9.
14 Texas Law Review [Vol. 89:XXXX
72. See LOVEMAN, supra note 46, at 54 (acknowledging newly formed Latin American nations‘
incorporation of rights and liberties from the Cádiz constitution and French Revolutionary ideals).
73. See JOHN A. HAWGOOD, MODERN CONSTITUTIONS SINCE 1787, at 49–58 (Fred B.
Rothman & Co. 1987) (1939) (comparing the Portuguese, French, Spanish, and Norwegian
constitutions of the era, and noting the limits on the monarch‘s powers in each); LOVEMAN, supra
note 46, at 40–45 (describing the limited role of the Spanish monarch under the Cádiz constitution).
2011] Latin American Presidentialism 15
A. Emergency Powers
A word is in order as to why we consider emergency powers to be
legislative in nature. First, periods of emergency rule generally allow for the
temporary delegation of considerable powers—including those normally
vested in the legislature—to the executive.74 The easier it is to declare a state
of emergency, the more likely it will be that the executive will predominate
and in some cases even usurp legislative authority strategically. Second, the
executive may be able to act without legislative authorization, as Ferejohn
and Pasquino recognized in their study distinguishing between constitutional
and legislative models of emergency powers.75 In their legislative model,
ordinary legislation facilitates emergency power, and so there is not a true
―regime of exception‖76 outside constitutional constraints.77 But much
depends on the specific assignment of powers to declare an emergency and
then to legislate during one.
The U.S. Constitution provides for relatively narrow emergency powers.
The relevant clause provides that ―[t]he Privilege of the Writ of Habeas
Corpus shall not be suspended, unless when in Cases of Rebellion or
Invasion the public Safety may require it.‖78 By contrast, Bolivia‘s 1851
emergency provision imagines broader powers for the president (―to assume
extraordinary faculties‖) but also requires the consent through countersigna-
ture of all ministers of state in order to establish emergency conditions.79
This is a model with a legal constraint, but it does not fit the legislative
model fully because the legislature has no involvement. According to
Loveman, it is precisely these sorts of expansive emergency provisions that
74. See John Ferejohn & Pasquale Pasquino, The Law of the Exception: A Typology of
Emergency Powers, 2 INT‘L J. CONST. L. 210, 217 (2004) (―The legislative model handles
emergencies by enacting ordinary statutes that delegate special and temporary powers to the
executive.‖); Mark Tushnet, The Political Constitution of Emergency Powers: Parliamentary and
Separation-of-Powers Regulation, 3 INT‘L J.L. CONTEXT 275, 275 (2007) (―‗Emergency powers‘
describes the expansion of governmental authority generally . . . , and the transfer of important ‗first
instance‘ lawmaking authority from legislatures to executive officials, in emergencies.‖).
75. Ferejohn & Pasquino, supra note 74, at 211–21.
76. See LOVEMAN, supra note 46, at 6 (establishing that many Latin American constitutions
contained provisions allowing the invocation of ―regimes of exception,‖ wherein executive
authority would be expanded, and constitutional protections, rights, and liberties would be
temporarily voided).
77. See Ferejohn & Pasquino, supra note 74, at 219 (―[B]ecause the legislature—the part of the
government closest to the people—actively delegates authority to the executive, the exercise of that
power is more constrained and legitimate and is even, indeed, amplified and made more efficient by
the fact that this exercise is supported by the legislature and, presumably, by the people.‖).
78. U.S. CONST. art. I, § 9, cl. 2.
79. CONSTITUCIÓN POLÍTICA DE LA REPÚBLICA DE BOLIVIA [BOL. CONST. 1851] Sept. 21,
1851, art. 76, cl. 26.
16 Texas Law Review [Vol. 89:XXXX
have threatened the political stability in the region almost from the
beginning.80
Nevertheless, emergency provisions vary in important ways across
Latin American constitutions. Appendix E, which reports the proportion of
constitutions that contain various emergency provisions, demonstrates some
of this variation. The variation spans at least three dimensions: (1) the iden-
tity of the actors (in addition to the executive) involved in the process of
declaring the existence of an emergency situation; (2) the conditions under
which emergency can be declared; and (3) limitations on the actions taken
under emergency conditions. The last two rows of Appendix E present,
respectively, the proportion of constitutions that require the legislature to
play some role in the process of declaring an emergency (i.e., it must
approve, or at least be consulted before, the declaration of an emergency),
and the proportion of constitutions that explicitly specify the conditions
under which an emergency may be declared. Across these dimensions, we
focus on four specific aspects of emergency provisions: (1) the participation
of the legislature in the emergency process; (2) the reference to internal secu-
rity reasons as a justification for emergency powers; (3) the explicit provision
for the suspension or restriction of rights during emergency; and (4) the pro-
hibition of constitutional amendments during emergency rule.
Several patterns are worth noting. The first is that almost all
presidential constitutions contain emergency provisions, whereas 73.6% of
parliamentary and 81.0% of semi-presidential constitutions do. Indeed,
every Latin American presidential constitution written since independence
contains a provision for the executive to declare an emergency, compared
with 94.4% of non-Latin American presidential constitutions.
The role of the legislature in the process of emergency declaration is
smaller in presidential constitutions than it is in either parliamentary or semi-
presidential ones. Only 19.0% of Latin American presidential constitutions
require that the legislature approve the state of emergency, and an additional
1.9% require that the legislature at least be consulted, whereas 43.5% of non-
Latin American presidential constitutions require some form of legislative
participation (approval or consultation). There is a trend toward increasing
the participation of legislatures in declaring emergencies, but this trend is
weaker in presidential constitutions in Latin America than elsewhere: of the
more recent (post-1979) constitutions, 36.8% of Latin American presidential
constitutions require legislative participation, while 55.9% of non-Latin
American, non-presidential constitutions do.
It is likely that the criteria stipulated in a constitution for identifying an
emergency tell us something about the flexibility of the executive‘s power
under these conditions. We can speculate that constitutions that limit
80. See LOVEMAN, supra note 46, at 6–9 (stating that while the clauses granting emergency
powers ―did not cause violence and dictatorship,‖ they ―are the constitutional foundation for tyranny
almost everywhere in Latin America today‖).
2011] Latin American Presidentialism 17
emergencies to natural disasters are less flexible than ones that allow for
emergencies for public security reasons. Further, we can speculate that con-
stitutions that allow for emergencies in response to internal security issues
are particularly broad, since they do not require an external trigger such as an
invasion. We find that reference to internal security reasons as a justification
for the state of emergency is more common in presidential than it is in par-
liamentary or semi-presidential constitutions; it is more common in Latin
American than in non-Latin American presidential constitutions; and
although it has become more common in all constitutions over the years, it is
considerably more common in Latin American presidential than in non-Latin
American, non-presidential constitutions. Emergency rule seems easier to
invoke in Latin America than it does elsewhere.
Presidential and parliamentary constitutions are equally likely to contain
an explicit provision allowing for the suspension or restriction of rights dur-
ing emergency rule. Among presidential constitutions, however, those in
Latin America are considerably more likely to allow for the suspension of
rights than those outside of Latin America. The proportion of Latin
American presidential constitutions with such a provision has hovered
around 90% in the post-World War II period.
Finally, only a small proportion of all constitutions explicitly forbid
legislative dissolution or constitutional amendments during emergency rule.
It is not surprising that this proportion is much smaller in presidential than in
parliamentary and semi-presidential constitutions, since assembly dissolution
by the executive is not a common item on the ordinary menu of presidential
powers outside assembly-confidence systems.81 Latin American presidential
constitutions, however, are about one-fifth as likely as non-Latin American
presidential constitutions to contain a provision prohibiting legislative
dissolution under emergency rule. We do not know if this distinction reflects
the fact that a Latin American presidency is particularly empowered vis-à-vis
the legislature (because it is free to dissolve the assembly) or disempowered
(because it is never allowed to dissolve the assembly and so the constitution
is silent about the rule during emergencies), but it is at least possible that the
former is the case. To summarize, Latin American presidential constitutions
are relatively less likely to require some form of legislative participation for
the activation of emergency powers; more likely to permit internal security
concerns as justifying the state of emergency; and more likely to explicitly
allow for the suspension or restriction of rights. This is largely consistent
with Loveman‘s claims.82 It seems unlikely, however, that the presence of
these provisions exhibits any causal relationship with the instability that has
81. See CHEIBUB, supra note 13, at 10 (remarking that the threat of dissolution is ―absent, by
design,‖ from presidential constitutions); cf. Ginsburg, Melton & Elkins, supra note 59, at 1816
(noting that in popular-election systems, the legislature and executive are constituted independently,
while in assembly-confidence systems, either branch can dissolve the other).
82. See supra note 80 and accompanying text.
18 Texas Law Review [Vol. 89:XXXX
B. Decree Powers
Executive decree powers give the executive the ability to issue binding
rules with the force of law and are an important feature of modern
governments,83 being found in about two-thirds of all constitutions. The
design of decree powers varies widely across cases, as illustrated in
Appendix F, and their rationale and distribution depends on the broader
political system: about 70% of both semi-presidential and presidential con-
stitutions provide such power for the executive, while only half of
parliamentary systems do so. In systems with fused governmental powers
(parliamentary and some semi-presidential constitutions), the decree power
for the executive is usually conceived as the exercise of delegated power
from the legislature. The legislature, therefore, is frequently designated as
the body that must approve an executive decree in those systems. In contrast,
only 27.0% of presidential constitutions designate the legislature as the
approving body of executive decree powers. Instead, such systems often
require that the executive approve the decree, meaning in practical terms the
cabinet in most cases. This is consistent with a conception of separation of
powers and the notion of a discrete realm of executive lawmaking.
In neither system is it the case that executives are unconstrained in their
ability to issue decrees. In fact, the difference between the three systems
almost disappears when we consider whether the constitution specifies that
some governmental body—be it the legislature or the cabinet—must approve
executive decrees. The numbers (not shown in the Appendix) are 68.3% for
presidential constitutions and 73.2% for both parliamentary and semi-
presidential constitutions.
In keeping with the logic of fused powers and delegated authority,
parliamentary and semi-presidential constitutions are twice as likely as
presidential constitutions to stipulate that, once issued, executive decrees are
immediately effective. Presidential constitutions are significantly more
likely than parliamentary and semi-presidential constitutions to require that
the approving body (the legislature or the cabinet or both) approve the decree
before it becomes effective. Thus, at least in this respect, the executive is
83. See John M. Carey & Matthew Soberg Shugart, Calling Out the Tanks or Filling Out the
Forms?, in EXECUTIVE DECREE AUTHORITY 1, 9, 15–19 (John M. Carey & Matthew Soberg
Shugart eds., 1998) (defining decree as ―the authority of the executive to establish law in lieu of
action by the assembly,‖ and discussing its appeal as a component of democratic government).
2011] Latin American Presidentialism 19
84. These numbers refer to the sum of the rows labeled ―Permanent, unless repealed‖ and
―Naturally expires, unless extended‖ in Appendix F.
20 Texas Law Review [Vol. 89:XXXX
85. In our sample, there are nine out of 444 (representing 1.99%) constitutions that do not
explicitly provide for a revision mechanism; two are presidential, six are parliamentary, and one is
semi-presidential. None of these is in Latin America.
86. See, e.g., Shugart & Mainwaring, Rethinking the Terms of the Debate, supra note 12, at 32
(observing that constitutional mechanisms for resolving these kinds of conflicts are of ―doubtful
democratic legitimacy‖).
87. See, e.g., id. at 30 (explaining that while most presidential systems with fixed executive
terms have provisions for impeachment, ―they offer less flexibility in crisis situations because
attempts to depose the president can easily endanger the regime itself‖).
2011] Latin American Presidentialism 21
88. See M. DE SECONDAT, BARON DE MONTESQUIEU, THE SPIRIT OF THE LAWS 162–63, 172
(J.V. Pritchard ed., Thomas Nugent trans., Fred B. Rothman & Co. 1991) (1914) (noting that the
legislative power enacts the laws and the executive carries out all functions of the state not reserved
to the judiciary, but that the executive ―has no other part in the legislative [power] than the privilege
of rejecting‖).
89. See Carl T. Bogus, The Battle for Separation of Powers in Rhode Island, 56 ADMIN. L.
REV. 77, 91 (2004) (―Montesquieu especially influenced the American Founders on the concept of
separation of powers.‖); Susanna Frederick Fischer, Playing Poohsticks with the British
Constitution? The Blair Government’s Proposal to Abolish the Lord Chancellor, 24 PENN ST. INT'L
L. REV. 257, 283 (2005) (―Montesquieu‘s views on the separation of powers are at least somewhat
familiar to most Americans, because his writings had such a profound influence on some of the
American Founders.‖); Ken I. Kersch, Justice Breyer’s Mandarin Liberty, 73 U. CHI. L. REV. 759,
780 (2006) (book review) (―[T]he Constitution, . . . and the American people, were fully committed
to government by elected representatives, an independent judiciary, [and] separation of powers
more generally . . . , thanks in large part to the influential writings of [the] liberal French thinker,
Montesquieu.‖). For a discussion of the influence of the American founders on other constitutional
drafters, see Bruce Ackerman, The New Separation of Powers, 113 HARV. L. REV. 633, 634–42
22 Texas Law Review [Vol. 89:XXXX
been argued that the rise of the so-called ―plebiscitary presidency‖ has
changed the structure of the office.91 Presidents are typically the only figures
elected by a national constituency and hence are more likely to reflect the
preferences of the median voter.92 Congress, in contrast, is seen as respond-
ing to a myriad of local interests, and hence it is not expected to produce
policies truly in the national interest.93 Furthermore, policy in Congress is
produced through a complex process of committees, vote trading, and nego-
tiation across houses, which makes it difficult to assign responsibility for any
particular policy. When a single individual holds responsibility, the public
clearly knows whom to blame or credit for policies. Executive lawmaking,
in this view, facilitates accountability.
In contrast with these functional accounts, some have asserted that the
concentration of authority in the presidency reflects a naked power grab.
This is the view associated with Loveman and others who argue for the con-
tinuing relevance of the caudillo tradition in Latin America.94 These scholars
emphasize the use of the emergency power by Latin American presidents.95
The emergency power, they show, has long been used to take power from the
legislature and leads to periods of executive tyranny.96 The assignment of
decree power to the executive, in this view, also comes at the expense of the
legislature, in that the executive can use that power not only with regard to
the technical details of delegated lawmaking, but also for setting the broad
outlines of policy.97
A full evaluation of these competing positions is beyond the scope of
this Article, but we lean toward the view that there is something quite func-
tional about the expansion of executive lawmaking authority. There are two
reasons for our view. First, we observe the increasing power of single
91. See Jide Nzelibe, The Fable of the Nationalist President and the Parochial Congress, 53
UCLA L. REV. 1217, 1224–31 (2006) (contrasting the modern plebiscitary vision of the presidency
with the early Federalist vision, and surveying the modern scholarship and judicial conclusions
regarding the structural implications of a plebiscitary presidency).
92. See INS v. Chadha, 462 U.S. 919, 948 (1983) (observing that the President brings a
―national‖ perspective to the legislative process); Elena Kagan, Presidential Administration, 114
HARV. L. REV. 2245, 2335 (2001) (―[B]ecause the President has a national constituency, he is likely
to consider, in setting the direction of administrative policy on an ongoing basis, the preferences of
the general public, rather than merely parochial interests.‖).
93. Chadha, 462 U.S. at 948 (citing Myers v. United States, 272 U.S. 52, 123 (1926)).
94. LOVEMAN, supra note 46, at 398; see also R.A. HUMPHREYS, TRADITION AND REVOLT IN
LATIN AMERICA AND OTHER ESSAYS 220 (1969) (―[T]he caudillo tradition survives. Political
creeds exist, and some of them are increasingly important.‖).
95. See, e.g., LOVEMAN, supra note 46, at 5–6 (―Latin American constitutions almost always
included provisions for ‗emergency powers,‘ or ‗extraordinary powers‘ . . . [that] might be exercised
by presidents . . . .‖).
96. See, e.g., id. at 6 (―[N]ormal constitutional protections were suspended, rights and liberties
were temporarily voided, and the government‘s authority was greatly expanded.‖).
97. See id. at 21 (―[O]rdinary government procedures for legislation, administration, and
judicial decision making may be replaced temporarily with special methods for making and
implementing public policy.‖).
24 Texas Law Review [Vol. 89:XXXX
VI. Conclusion
We have analyzed the formal features of executive power in Latin
America, a region long understood to be one amenable to strong executive
rule. We have demonstrated that, although the presidency was inspired by
the American model, other models were equally influential in structuring the
precise contours of executive and legislative power in the region. We have
also seen increasing convergence within the region along important dimen-
sions of executive–legislative relations. We can thus speak of a Latin
American model of presidential power that includes a powerful role in legis-
lation as well as extensive emergency rule. This distinguishes the Latin
American presidency from those in other regions of the world.
Our analysis has several implications for the study of comparative law
and politics. First, it calls attention to geography as an important predictor of
constitutional design. Second, our analysis emphasizes change rather than
continuity and convergence over time. This approach contrasts with the
recent emphasis in comparative law on ―legal origins‖ as determinants of
98. See Thomas Poguntke & Paul Webb, The Presidentialization of Politics in Democratic
Societies: A Framework for Analysis, in THE PRESIDENTIALIZATION OF POLITICS 1, 5–6 (Thomas
Poguntke & Paul Webb eds., 2005) (explaining that the degree to which presidentialization occurs
in any system, including parliamentary ones, depends on a range of factors including ―changes in
the social structure and the media system‖).
2011] Latin American Presidentialism 25
99. Cf. generally Rafael La Porta et al., Law and Finance, 106 J. POL. ECON. 1113 (1998)
(examining the origins of legal rules covering the protection of corporate shareholders and creditors
in forty-nine countries, as well as the quality of their enforcement).
100. See id. at 1118 (―When the Spanish and Portuguese empires in Latin America dissolved in
the nineteenth century, it was mainly the French civil law that the lawmakers of the new nations
looked to for inspiration.‖).
26
Executive can freely dismiss the cabinet 89.5 82.5 91.3 96.6 90.6 95.0
Executive can freely appoint the cabinet 95.3 95.2 100.0 96.6 93.8 85.0
Executive can issue pardons 72.1 63.5 78.3 82.8 78.1 60.0
Executive has veto power 89.5 92.1 82.6 79.3 96.9 100.0
Legislature has oversight powers over the executive 90.0 77.8 95.7 93.1 96.9 100.0
Legislature can remove individual ministers 53.7 61.9 34.8 55.2 62.5 55.0
Legislature can override executive veto 85.8 88.9 78.3 75.9 93.8 95.0
Executive has emergency power 99.0 98.4 100.0 96.6 100.0 100.0
Executive can initiate constitutional amendments 23.7 7.9 4.4 27.6 37.5 90.0
Executive can initiate ordinary legislation 69.0 52.4 65.2 69.0 87.5 100.0
Executive can propose budget 42.1 19.1 21.7 51.7 81.3 85.0
Executive has decree powers 77.3 63.5 71.7 100.0 81.3 95.0
Appendix E. Regulation of Emergency Provisions in Constitutions that Grant Emergency Powers to the Executive, 1789−2007
Presidential Latin American Presidential Non-Latin American, Non-Presidential
Semi- Pre- 1870− 1919− 1946− Post- Pre- 1870− 1919− 1946− Post-
All Presidential Parliamentary presidential LA Non-LA 1870 1918 1945 1979 1979 1870 1918 1945 1979 1979
Constitutions with emergency provisions 87.2 98.0 73.6 81.0 100.0 94.4 100.0 100.0 100.0 100.0 100.0 91.7 90.0 65.0 75.3 83.8
Legislature cannot be dissolved 6.6 4.5 10.1 19.1 1.9 9.4 2.4 0.0 1.8 3.7 0.0 0.0 0.0 0.0 12.1 11.8
Constitution cannot be amended 1.8 1.2 2.3 7.4 1.3 1.2 0.0 0.0 0.0 3.7 5.3 0.0 0.0 0.0 2.6 4.3
No restrictions are imposed 4.5 5.8 3.4 1.5 5.7 5.9 7.3 6.8 6.3 0.0 10.5 3.0 5.6 1.6 6.9 1.1
Legislature plays a role 32.6 28.8 38.2 58.8 20.9 43.5 9.8 20.5 22.2 25.9 36.8 12.1 11.1 38.5 31.9 55.9
Specification of conditions 65.7 77.8 62.9 66.2 82.9 68.2 75.6 84.1 81.5 85.2 94.7 27.3 50.0 50.0 54.3 66.7
Note: Entries represent the percentage of constitutions in each category (column) that provide for a given feature (row).
31
32
Appendix F. Regulation of Decree Powers in Constitutions that Grant Decree Powers to the Executive, 1789−2007
Presidential Latin American Presidential Non-LA Pres.
Semi- Pre- 1870− 1919− 1946− Post- 1946− Post-
All Presidential Parliamentary presidential LA Non-LA 1870 1918 1945 1979 1979 1979 1979
Constitutions with executive decree powers 65.4 70.2 52.1 71.4 78.5 55.6 63.4 70.5 100.0 81.5 94.7 61.5 51.3
Decree Implementation
Effective immediately once issued 17.3 17.8 41.3 41.7 5.7 48.0 0.0 3.2 0.0 18.2 11.1 41.7 60.0
Effective following a specified period 1.1 1.7 0.0 3.3 0.8 4.0 0.0 0.0 0.0 0.0 5.6 4.2 5.0
during which an approving body can repeal it
Effective only after approval from the approving body 19.2 37.4 23.8 18.3 46.8 14.0 61.5 41.9 51.9 40.9 33.3 16.7 5.0
Not specified 27.5 37.9 38.1 43.3 42.7 26.0 34.6 51.6 44.4 36.4 44.4 25.0 30.0
Decree validity
Permanent, unless repealed 7.3 8.0 12.7 10.0 4.0 18.0 3.9 3.2 0.0 4.6 11.1 16.7 20.0
Naturally expires, unless extended 10.5 10.3 23.8 18.3 3.2 28.0 0.0 0.0 0.0 18.2 0.0 25.0 35.0
Not specified 47.0 75.3 46.0 56.7 87.9 44.0 96.2 90.3 92.6 77.3 77.8 45.8 40.0
Legislature 24.8 27.0 61.9 51.7 19.4 46.0 7.7 9.7 22.2 36.4 27.8 37.5 60.0
Not specified 20.6 26.4 30.2 33.3 26.6 26.0 19.2 29.0 29.6 18.2 38.9 25.0 30.0
In case of the legislature's failure to pass a proposal, the budget defaults to:
Executive's proposal 9.3 10.7 6.3 9.5 7.6 16.2 0.0 0.0 4.6 23.1 11.1 15.2 17.9
Executive's proposal, if
proposed prior to failure 12.8 11.4 11.1 27.3 12.9 5.6 0.0 0.0 7.1 27.3 12.5 0.0 14.3
Previous year's budget 33.2 32.1 28.4 41.9 30.3 35.3 5.3 32.4 31.8 42.3 33.3 45.5 28.6
Other 14.9 9.1 23.2 18.9 4.2 17.7 0.0 0.0 9.1 7.1 5.6 24.2 10.7
Note: Entries represent the percentage of constitutions in each category (column) that provide for a given feature (row)
33
Readers with comments may address them to:
346. Rosalind Dixon and Richard Holden, Constitutional Amendment Rules: The
Denominator Problem, May 2011
347. Rosalind Dixon, Constitutional Amendment Rules: A Comparative Perspective, May
2011
348. Rosalind Dixon, Weak-Form Judicial Review and American Exceptionalism, May 2011
349. Rosalind Dixon, Transnational Constitutionalism and Unconstitutional Constitutional
Amendments, May 2011
350. Adam B. Cox and Richard T. Holden, Reconsidering Racial and Partisan
Gerrymandering, May 2011
351. Brian Leiter, The Circumstances of Civility, May 2011
352. Brian Leiter, Naturalized Jurisprucence and American Legal Realism Revisited, May
2011
353. Lee Anne Fennell, Property and Precaution, June 2011
354. Alon Harel and Ariel Porat, Commensurability and Agency: Two Yet-to-Be-Met
Challenges for Law and Economics, June 2011
355. Bernard E. Harcourt, Radical Throught from Marx, Nietzsche, and Freud, through
Foucault, to the Present: Comments on Steven Lukes’ “In Defense of False
Consciousness,” June 2011
356. Alison L. LaCroix, Rhetoric and Reality in Early American Legal History: A Reply to
Gordon Wood, July 2011
357. Martha C. Nussbaum, Teaching Patriotism: Love and Critical Reform, July 2011
358. Shai Dothan, Judicial Tactics in the European Court of Human Rights, August 2011
359. Jonathan S. Masur and Eric A. Posner, Regulation, Unemployment, and Cost-Benefit
Analysis, August, 2011
360. Adam B. Cox and Eric A. Posner, Delegation in Immigration Law, September 2011
361. José Antonio Cheibub, Zahcary Elkins, and Tom Ginsburg, Latin American
Presidentialism in Comparative and Historical Perspective, September 2011